Transmit Energy Focusing for Parameter Estimation in Slow-time Transmit Beamspace L-shaped MIMO Radar

Abstract

We present a novel slow-time transmit beamspace (TB) multiple-input multiple-output (MIMO) technique for L-shaped array radar with uniform linear subarrays to estimate target parameters including 2-dimensional (2-D) directions of arrival (DOA) and unambiguous velocity. Doppler division multiple access (DDMA) approach, as a type of slow-time waveform achieving waveform orthogonality across multiple pulses within a coherent processing interval, disperses the transmit energy over the entire spatial region, suffering from beam-shape loss. Moreover, Doppler spectrum division, which is necessary for transmit channel separation prior to parameter estimation, leads to the loss of crucial information for velocity disambiguation. To optimize transmit energy distribution, slow-time TB technique is proposed to focus the energy within a desired spatial region. Unlike DDMA approach, slow-time TB technique divides the entire Doppler spectrum into more subbands than the number of transmit antenna elements to narrow down the beam mainlobe intervals between adjacent beams formed by DDMA modulation vectors. As a result, more beams are incorporated into the region of interest, and slow-time TB radar can direct transmit energy to the region of interest by properly selecting the DDMA modulation vectors whose beams are directed there. To resolve velocity ambiguity, tensor signal modeling, by storing measurements in a tensor without Doppler spectrum division, is used. Parameter estimation is then addressed using canonical polyadic decomposition (CPD), and the performance of slow-time TB L-shaped MIMO radar is shown to be improved as compared to DDMA MIMO techniques. Simulations are conducted to validate the proposed method.